Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    specific named locations associated with tragic or significant events

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and formatting related to written communication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     nodd
    -0.82
     earthqu
    -0.71
     spitting
    -0.70
     carbohyd
    -0.68
     petrol
    -0.66
     princ
    -0.66
     frontline
    -0.66
     sensit
    -0.65
     favour
    -0.65
     tiss
    -0.64
    POSITIVE LOGITS
    .?
    1.48
    .-
    1.37
    .:
    1.34
     .
    1.33
    ._
    1.28
    -.
    1.25
    .,
    1.23
    .(
    1.23
    .
    1.22
    .;
    1.18
    Act Density 0.159%

    No Known Activations